- Email: [email protected]
Independent Functioning of the Two Cerebral Hemispheres for Recognizing Bilaterally Presented Tachistoscopic Visual-Half-Field Stimuli
INDEPENDENT FUNCTIONING OF THE TWO CEREBRAL HEMISPHERES FOR RECOGNIZING BILATERALLY PRESENTED TACHISTOSCOPIC VISUAL-HALF-FIELD STIMULI David Hines (Department of Behavioral Science, The Milton S. Hershey Medical Center of The Pennsylvania State University)
INTRODUCTION
Words are recognized more accurately from the right visual half-field (VHF) under conditions of tachistoscopic presentation. At least part of this superiority is often attributed to the more direct pathway between the right VHF and the language centers in the left hemisphere (Kimura, 1966; Hines, 1972b). This effect has usually been reported using stimuli presented unilaterally to only one VHF, on any given trial. Recently, however, McKeever (1971) has shown that when words are presented bilaterally to both VHF's, with fixation control using a center digit, a significantly greater right VHF superiority is produced. This effect is quite large. With unilateral presentation McKeever (1971) found a right VHF superiority of about 1.5 to 1; with bilateral presentation he found a ratio of about 5 to 1. The 5 to 1 ratio has also been reported by McKeever and Huling (1971a) and by Hines ( 1972a). Several hypothesis to explain the larger right VHF superiority with bilateral stimulation have been proposed and tested, but none have satisfactorily accounted for all the experimental data. McKeever (1971) tested the effect of order of report on the bilateral VHF asymmetry but he found the large right VHF asymmetry was maintained even when subjects were instructed to report the left VHF first. McKeever and Huling (1971b) tested a "temporal lag" hypothesis to explain the larger VHF asymmetry. They suggested that the right VHF word arrived first at the language areas in the left hemisphere, since it did not have to cross the callosum. They also hypothesized that the left VHF word arrived in the left hemisphere, via the callosum, while the left hemisphere language center already was "engaged in processing or recovering from processing" the right VHF word. McKeever and Huling tested this hypothesis by presenting the left VHF word either 6 msec., 10 msec., or 20 msec. prior to the right VHF word. The 6 msec. time was thought to Cortex (1975) 11, 132-143.
Independent functioning of. cerebral hemispheres
133
be roughly the time necessary to cross the callosum. However, they found that earlier presentation of the left WIF stimulus did not result in a decrease in right VHF superiority. Hines (1972a) and White (1973) suggested that the center digit might, in itself, alter fixation. Hines suggested that after reading the center digits, .. subjects may simply continue reading to the right." However, McKeever, Suberi and Van Deventer (1972), Olsen (1972) and Kershner and Jeng (1972) demonstrated that other methods of fixation control which do not use a center digit also produce the large right VHF superiority. The present experiment tests two additional hypotheses to explain the increased right VHF superiority with bilateral stimulation. The first hypothesis is that the increase is due to subjects directing attention toward the more easily recognizable right VHF word, and tending to ignore the left VHF word. This hypothesis will be tested in two ways. First, we will evaluate whether accurate report of words from the right VHF is associated with decreased accuracy from the left VHF on any given trial. Second. we will test whether the large right VHF superiority is obtained when words are presented in one VHF and nonverbal stimuli in the opposite VHF. With bilateral presentation of words, very few words are recognized from the left VHF. However, nonverbal shapes show much less asymmetry with bilateral presentation and fixation control. Hines (1972a) found that subjects reported about 25 % of the shapes from the left VHF with bilateral presentation but less than 10% of the words. If subjects can also recognize about 25% of the shapes from the left VHF when they are paired with words, without lowering recall of words from the right VHF, then it will suggest the left VHF "deficit" is not due to attentional factors, but rather is specific for verbal stimuli. If, however, subjects cannot recognize nonverbal stimuli from the left VHF, without decreasing recognition of words from the right VHF, it will suggest that attentional factors may be important. The second hypothesis to be tested is based on Dimond's (1972) "channel" theory of visual perception. Dimond postulates that stimuli to each VHF are received and analyzed by the contralateral hemisphere. Thus any stimuli presented to the right VHF will be initially processed by the left hemisphere, while stimuli to the left VHF will be initially recognized by the right hemisphere. The recognition accuracy from each VHF depends on the processing efficiency of its contralateral hemisphere for any particular type of stimuli. Dimond has also proposed that while bilaterally presented visual stimuli are received and recognized separately by the two hemispheres working as "independent channels," the stimuli may interfere with each other if they have to compete with each other for access to the expressive language centers located in the left hemisphere. Dimond's channel theory can be extended to account for the increased right VHF superiority with bilateral stimulation
134
D. Hines
as due to interference between the hemispheres since both must use the same expressive language areas. Thus the left hemisphere recognition channel may have preferential access to the language areas of the left hemisphere. In the present study, we would predict that subjects should show suppression of words from the left VHF due to interference only when words are presented bilaterally to each VHF. When the words are presented bilaterally, they may compete for the language mechanisms in the left hemisphere causing a loss of the word processed by the right hemisphere. However, when only a recognition of a nonverbal stimulus is required of the left hemisphere, then the verbal information from the right hemisphere should not be suppressed since the left hemisphere is not specialized for nonverbal recognition (Kimura, 1966; Levy, Trevarthen and Sperry, 1972). Thus when words are paired with nonverbal stimuli they should show a similar asymmetry to words presented unilaterally on successive trials. Three types of stimuli will be used in the present experiment: words, nonverbal shapes, and pictures of faces. A center digit will be used to control fixation. In all conditions, the stimuli will be presented bilaterally. In three conditions the same type of stimulus will be presented to each VHF: words to both VHF's, shapes to both VHF's, and faces to both VHF's. In two conditions a word will appear in one VHF and a nonverbal stimuli in the opposite VHF: words paired with shapes and words paired with fates. In these two conditions, the VHF in which the verbal and nonverbal stimuli appear will be randomly alternated between trials. The'attentional theory suggests that correct recognition from one VHF should be correlated with decreased recognition from the opposite VHF in all conditions. It also suggests that superior recall of verbal stimuli from the right VHF should decrease recall of verbal and nonverbal stimuli from the ignored left VHF. Dimond's theory suggests that stimuli should interfere only when they compete for a common response mechanism. Thus words should be reported more accurately when paired with nonverbal stimuli, particularly the words from the left VHF. MATERIAL AND METHOD
Subjects
Sixty three students and employees at the Hershey Medical Center of The Penn State University served as paid subjects. All subjects were naive as to the purpose of the experiment. Partial data from 18 subjects in the "shape-pairs" control condition have been reported previously (Hines, 1972a). Four additional subjects failed to make any correct responses and were excluded from the study. There were five experimental conditions in this study: words paired with words; words paited with. shapes; words paired with faces; shapes paired
Independent functioning of cerebral hemispheres
135
with shapes; and faces paired with faces. Fifteen subjects were randomly assigned to the words paired with shapes condition; fifteen subjects were assigned to the words paired with faces condition; fifteen subjects were assigned to the faces paired with faces condition and eighteen subjects were assigned to the shapes paired with shapes condition. Fifteen subjects received the words paired with words condition. Nine of these subjects had previously received the shapes paired with shapes condition and six the faces paired with faces condition. Equipment and stimulus preparation
The stimuli were presented using a Lafayette Model V-I constant illumination tachistoscope. The stimulus cards were prepared using index cards trimmed to fit the V-I tachistoscope. All cards had a number from one to nine typed at the central fixation point. Three types of stimuli were presented in the VHF's: words, shapes and pictures of faces. The words and central digits were typed on an IBM electric typewriter with standard pica type. The words were typed in all capital letters with one space between the letters and four spaces between the word and center digit. The inner edges of each word was 1.6 and the outer edge 4.1 degrees of visual arc from the center point of fixation at a viewing distance of 15 inches. Ten four-letter words were presented in the three word conditions (words paired with words, words paired with faces, words paired with shapes): CAKE, LANE, DOVE, MASK, EPIC, GOLD , HARE, FARM, POST and BEAR. All words appeared twice in each VHF in each of the three word conditions. The words used and stimulus preparation are identical to McKeever and Huling (1971b). The word and word pair stimuli were exposed for 20 msec. The "shapes" were prepared by the Medical Illustration Department of Penn State. They consist of inkblot-like shapes drawn to avoid resemblance to any familiar objects. On the stimulus cards, the inner edge of each shape was 1.9 degrees and the outer edge was 4.0 degrees of visual arc from fixation. Twenty of the shapes and a mirror image of each of the twenty were used as the stimuli in all shape conditions. Each of the shapes appeared once in one VHF and its mirror image appeared once in the opposite VHF in both the "shape and shape pairs" and "shape and word pairs" conditions. The mirror images were used to avoid any left-right recognition asymmetries inherent in the shapes themselves as suggested by Kimura (1966). For each trial an answer card containing six possible choices wasaIso prepared. Illustrative stimulus and answer cards have been published (Hines 1972a). The answer cards for each trial was placed adjacent to the tachistoscope. Subjects viewed the stimulus in the tachistoscope, then looked down at the answer card and pointed to the shape they felt they had seen. The shape and shape pairs and the shape and word pairs stimuli were exposed for 25 msec. A pilot study suggested that this was the shortest exposure at which most subjects could accurately recognize some shapes. The face pictures were obtained from photographs of medical students taken by the Department of Medical Illustrations at Penn State. The photographs were 3/4 inch by 7/8 inch in size and contained the full face of one student. The photographs were usually homogeneous with respect to background. darity, contrast, and face size. Twenty photographs and a mirror image of each photograph were utilized as stimuli in each face condition. There was 1.9 degrees of arc from the center to the inner edge of each photograph on the stimulus cards, and 4.5 degrees from fixation to the outer edge of each photograph. Each of the faces appeared once in one VHF on one trial and its mirror image appeared in the
136
D. Hines
opposite VHF on another trial. This procedure controls for any left or right bias inherent in the stimuli (Kimura, 1966). Forty answer cards were prepared for the "face and word" pairs conditions. Each card contained the correct face plus three alternatives on a 3" X 5" plain index card. The three alternatives were matched to the correct choice for sex, hair length, facial hair, presence or absence of glasses, dominant expression, background darkness, and orientation of the face. A pilot study suggested the latter two characteristics were particularly easy to discriminate in tachistoscopic exposure. The purpose of this matching was to eliminate differences between faces which could be easily verbalized. Twenty answer cards were also prepared for the face and face pairs condition. Those answer cards contained two correct choices plus three alternatives for each correct choice. The same alternatives for each face were used as in the face and word pairs answer cards. The appropriate card for each trial was placed adjacent to the tachistoscope. Subjects viewed the stimulus cards in the tachistoscope. then looked down at the answer cards and pointed to the face(s) they felt they had just seen. The face and face pair and the face and word pair stimuli were exposed for 30 msec. Pilot data suggested that this was the shortest exposure at which most subjects could accurately recognize some faces.
Procedure Sl,lbjects began each trial by viewing a card with 6 lines radiating from an open circle. The center circle was just large enough to "frame" the center digit. None of the 6 radiating lines was horizontal and they did not overlap with any VHF stimuli. Prior to each trial the subjects were alerted to ensure fixation onto the center circle. The stimulus card was then exposed, followed immediately by the return of the fixation card. Subjects in the word and shape pairs condition were instructed to fixate the center circle until the stimulus was exposed, to first report the center digit and then to name the word and select the shape from the answer card. There were forty trials for each subject. On each trial a word appeared in one VHF and a shape in the other. The VHF in which the word and shape appeared was randomly alternated between the trials. Subjects in the word and face pairs condition were instructed to fixate the center circle until the stimulus appeared, to first report the center digit and then to name the word and select the correct face from the answer card. There were forty trials for each subject. On each trial a word appeared in one VHF and a face in the other VHF. The VHF in which the word and face appeared was randomly alternated between trials. Subjects administered the face and face pairs condition were instructed to fixate the center circle until the stimulus appeared, to first report the center digit and then to select the two faces from the answer card. There were twenty trials for each subject. The faces in the two half-fields were dissimilar on at least one significant aspect of appearance. Subjects in the shape and shape pairs condition were instructed to fixate the center circle until the stimuli were flashed, then to report the number presented at fixation, and last to select the shape they had seen from an answer card. There were twenty trials for each subject. Subjects in the word and word pairs condition were instructed to fixate the center circle until the stimuli were exposed, report the number at fixation and
Independent functioning of cerebral hemispheres
137
then any words they recognized. This condition was essentially a bilateral replication of McKeever and Ruling (1971a). RESULTS
The mean correct responses for all conditions are shown in Table I. The data were first analyzed by analysis of variance to determine whether VHF asymmetry varied according to stimulus pairings for each type of stimuli. TABLE I
Mean Correct Score by Visual Half-Field for Each Condition Stimulus pairs
Mean correct recognition Left VHF
Right VHF
Mean Correct Words Words and Words Words and Shapes Words and Faces Shapes and Shapes Shapes and Words
2.4 4.7
9.9 7.8 10.3
Mean Correct Shapes 5.1 5.1
7.2 6.3
2.8
Mean Correct Faces Faces and Faces Faces and Words
5.6 5.5
4.8 6.2
The maximum possible score for each visual half-field is 20.
The mean correct responses for words was analyzed by a mixed analysis of variance with stimulus pairing as a between groups factor and the VHF's and interaction as within group variables. Significantly more words. were recalled from the right VHF in all stimulus pairings (F = 107.21; df = 1, 42; P < .001). However, the stimulus pairings did not significantly affect either overall recall (F = 1.28; df = 2, 42) or VHF asymmetry (F = 1.63; df = 2, 42). The mean correct response for shapes was also analyzed by a mixed analysis of variance. The right VHF was superior in overall recognition (F = 4.75; df = 1, 31; P < .05). Again, however, stimulus pairings did not alter either overall recall (F = .15; df = 1,31) or VHF asymmetry (F = .10; df = 1, 31). Mean correct recognition for faces was also analyzed using a mixed analysis of variance. The faces showed no significant difference between VHF's in overall recognition (F = .21; df = 1, 28). As with words and
D. Hines
138
shapes, stimulus pamngs did not alter either total recognitions (F = 1.45; df = 1, 28) or VHF asymmetry (F = .84; df = 1, 28). These analysis indicate that the different stimulus pairings did not significantly alter VHF asymmetry or overall recognition for either the words, shapes, or faces. The data were then analyzed to determine whether the verbal and nonverbal stimuli showed significantly different VHF asymmetries when paired together. Data from the shapes and words paired condition were analyzed by analysis of variance, with both VHF and stimulus type within group variables. The words showed a significantly greater right VHF superiority than did the shapes, resulting in a significant VHF X stimuli interaction W = 17.63; df = 1, 14; P < .001). Data from the word and face pair condition were similarly analyzed. Again the words showed a right VHF superiority significantly greater than the faces (F = 16.00; df = 1, 14; P < .005). Thus while stimulus pairing did not significantly alter VHF asymmetry, both types of nonverbal stimuli did show a significantly different ViHF asymmetry from the verbal stimuli. The data were finally analyzed to determine whether correct recall from one VHF was associated with incorrect recall from the opposite VHF on the same trial. For each subject, the number of trials in which both VHF's were reported correctly was compared with the number expected by chance simply on the basis of total recall in each VHF. For example, if in 20 trials a subject had 8 correct responses from one VHF and 10 correct responses from the other VHF, by chance 10 X 8/20 or 4 trials should have occurred in which both responses were correct. The mean number of trials per subject in which both VHF's were reported correctly.and the mean number of trials per subject expected by chance are presented for each stimulus condition in Table II. Differences between the actual and chance distribution were tested by the Wilcoxon signed-rank test. TABLE II
Mean Trials in which Stimuli are Correctly Reported from both Visual Hall-Field (VHF) as Expected by Chance Type of stimuli Right VHF
Left VHF
Words Words Shapes Shapes Words Faces Faces
Words Shapes Words Shapes Faces Words Faces
* p.OI ** p.02
Both stimuli reported correctly Change
Observed
1.87 .97 2.04 2.08 1.39 3.04 1.27
1.86 1.27 2.00 1.00* . 1.20 ·3.20 .73**
Independent functioning of cerebral hemispheres
139
In two conditions, significantly fewer double correct responses occurred than would be expected by chance. When shapes were paired with shapes less than half the expected double correct responses occurred (T = 16, N = 18, P < .01). When faces were paired with faces there were also less trials when faces from both VHF's were correctly reported than would be expected by chance. (T = 19.5, N = 15, P < .02). Significantly more double responses occurred when the nonverbal stimuli were presented to the left VHF and the words to the right VHF, than when words were presented to the left VHF and shapes or faces to the right VHF (shapes: T = 3, N = 8, p < .05; faces: T = 0, N = 11, P < .01). However this reflected higher overall recognition when the verbal stimulus was presented to the right VHF and the nonverbal stimulus to the left VHF and the number of double responses for each of the verbal-nonverbal pairing was not different from chance expectancy. DISCUSSION
The results of this experiment show that the overall recogmtlon and VHF asymmetry depended on the type of stimulus and not on the stimulus paidngs. The words showed a large right VHF superiority whether paired with words, shapes, or faces. The shapes showed a small but statistically reliable right VHF superiority whether paired with words or shapes. The faces showed no significant differences between VHF's whether paired with words or faces. The large right VHF superiority for words is typical of this bilateral recognition paradigm. The right VHF superiority for shapes is not typically found with unilaterally presented nonverbal stimuli. In 25 experiments summarized by White (1972), nonverbal stimuli were recognized more accurately from the left VHF in 7 experiments, no difference was found in 15 experiments, and a right WfF superiority was reported in only 3 experiments. The lack of asymmetry for faces is consistent with the unilateral findings. These results indicate that the large right VHF superiority for bilaterally presented words is not a result of attending the right VHF and ignoring the left VHF. If subjects preferentially attend one VHF, then correct recognition from that VHF should be correlated with impaired recognition from the opposite VHF. However the recognition accuracy from one VHF did not affect recognition accuracy from the opposite VHF on the same trial in any of the word conditions . .In addition, if the left VHF is simply being ignored, then all types of stimuli should be recognized with similar impaired accuracy when words are being recoghized accurately from the left VHF. However, the nonverbal stimuli could be recognized accurately from the left VHF while words were recognized with unimpaired accuracy from the right VHF. Finally,
140
D. Hines
even though words were recognized with low accuracy from the left VHF in all conditions, the high accuracy from the right VHF was only observed in the words and word pairs conditions. However, if subjects were preferentially attending the right VHF, it should increase recognition for all right VHF stimuli. The latter effect is particularly noticeable in the word and shape pairs condition. The data are also inconsistent with the hypothesis that the bilateral verbal stimuli interfere with each other because both use the same language' response mechanism. If this hypothesis were correct, then the large VHF asymmetry for words should only be observed in the word and word pairs condition, when both words must be reported using the left hemisphere expressive language areas. However, a large right VHF superiority for words was observed in all conditions. The different stimulus pairings did not significantly affect either VHF asymmetry or overall recognition for words. Also, if the interference hypothesis were correct, then correct recognitions from one VHF should be correlated with incorrect recognitions from the opposite VHF on the same trial in the word and word pairs condition. However, VHF recognition on any given trial were not correlated in the word and word pairs condition. The most interesting positive finding was the interference noted with the two bilateral nonverbal conditions. When a nonverbal stimulus was recognized from either VHF, it significantly lowered the probability of correct recognition of a nonverbal stimulus from the opposite VHF. This was found in both the face and face pairs and the shape and shape pairs conditions. This finding suggests that subjects are relatively limited in their ability to keep unfamiliar nonverbal stimuli in short term memory, perhaps to only one stimulus at a time. While the data from this experiment do not support Dimond's interference hypothesis for verbal stimuli, they are otherwise consisted with channel theory. Each type of stimuli displayed the same VHF asymmetry regardless of stimulus pairing. This is consistent with the theory that the response asymmetry is due to differences between the two hemispheres in recognizing different types of stimuli. A proposed new
selective channel theory
The results of this experiment are most consistent with the theory that the two hemispheres function as separate channels for recognizing visual stimuli only under conditions of bilateral tachistoscopic VHF stimulation with fixation control. In this paradigm each hemisphere serves as an independent channel for recognizing stimuli from its contralateral VHF. VHF asymmetries under this condition reflect actual differences between the hemispheres for recognizing different types of stimuli. Thus the large right
Independent functioning of cerebral hemispheres
141
VHF superiority with bilateral presentation reflects the superior verbal recognition ability of the left hemisphere. It is also hypothesized that the two hemispheres do not function as separate channels under unilateral VHF tachistoscopic presentation. Rather, with unilateral VHF presentation both hemispheres contribute to recognizing incoming stimuli, with the primary role played by the hemispheres specialized for processing any given stimulus. Thus VHF asymmetry for tachistoscopically presented unilateral stimuli do not directly reflect differences between the two hemispheres in recognition ability. Rather it is due to information loss in transmission to the most efficient hemisphere when stimuli must cross the callosum. Thus the degree of VHF asymmetry with unilateral stimuli reflects the extent to which they are vulnerable to a slight loss of fidelity. Small complex stimuli (such as words) which are presented at rapid rates should be most affected. Larger stimuli presented at slower rates should be less affected and show less asymmetry. Selective channel theory thus suggests that differences in VHF asymmetry for words between the unilateral and bilateral conditions reflect two distinct and different mechanisms. The large right VHF superiority for words in the bilateral condition directly reflects the superior ability of the left hemisphere to recognize verbal stimuli. The smaller right VHF asymmetry with unilateral stimulation reflects a loss of information in transmission to the left hemisphere across the callosum. Selective channel theory finally proposes that the two hemispheres work as independent channels only when discrete tachistoscopic stimuli are presented bilaterally to each VHF, with positive fixation control. Under other bilateral presentation conditions, verbal stimuli are recognized sequentially from left to right with the recognition performed primarily by the left hemisphere. This, of course, is how English is normally read. The left to right order of recognition has also been found when letter spans are presented tachistoscopically across fixation to both VHF's (Heron, 1957; Bryden, 1960). Subjects report that they actually attend each letter from left to right (Heron, 1957). However, when fixation is controlled by a center digit, the left-to-right scan is no longer possible. Under these conditions, the two hemispheres work as independent channels to analyze information from their contralateral VHF.
SUMMARY
Bilateral tachistoscopic presentation of verbal stimuli produces a significantly larger right visual half-field (VHF) superiority than does unilateral presentation, when fixation · is controlled by a center digit. This experiment tested whether the increased asymmetry was due to (a) subjects attending the right VHF and ignoring the left VHF or (b) interference between the hemispheres due to competition for the left hemisphere language areas.
142
D. Hines
Words, shapes, and pictures of faces were presented bilaterally to each VHF, with fixation controlled by a center digit. In three conditions, the same type of stimuli was presented in each VHF (e.g., a word in both the left and right VHF). In two conditions words were presented to one VHF and nonverbal stimuli to the opposite VHF (words paired with words and words paired with faces). It was found that the stimulus pairings did not affect VHF asymmetry for any stimulus. Words showed a large right VHF superiority in all conditions. Shapes showed a significantly smaller right VHF superiority in all conditions. Faces showed no VHF asymmetry in any condition. It was concluded that attentional factors were not important ~ince shapes or faces could be recognized accurately from the left VHF without lowering verbal recognition from the right VHF. Thus the low recognition accuracy from the left VHF is specific for verbal stimuli rather than attentional. The interference hypothesis was also not supported since all the right VHF stimuli (words, shapes, or faces} were associated with low recognition of words from the left VHF. It was suggested instead that VHF asymmetry under unilateral and bilateral presentation reflect two different mechanisms. Under conditions of unilateral presentation, VHF asymmetries are caused by loss of information when any given stimulus must cross the callosum to reach the hemisphere specialized for its processing. However, with bilateral VHF presentation and fixation control, the two hemispheres act as independent channels for information processing. Under this condition, each hemisphere recognized the stimulus from its contralateral VHF. Thus the large .right VHF superiority for words with bilateral presentation reflects the superior ability of the .left hemisphere for verbal recognition.
REFERENCES BRYDEN, M.P. (1960) Tachistoscopic recognition of non-alphabetical material, "Canad. J. Psycho!.," 14, 78-86. DIMOND, S. (1972) The Double Brain, Williams and Williams, Baltimore. HERON, W. (1957) Perception as a function of retinal focus and attention, "Am. J. Psycho!.," 70, 38-48. HINES, D. (1972a) Bilateral tachistoscopic recognition of verbal and nonverbal stimuli, "Cortex," 8, 315-322. . ' - (1972b) A brief reply to McKeever, Suberi and Van Deventer's comment on "Bilateral tachistoscopic recognition of verbal and .nonverbal stimuli," "Cortex," 8, 480-482. KERSHNER, J.R., and JENG, A.G. (1972) Duel functional hemisphere asymmetry in visual perception: Effect of ocular dominance and postexposure processes, "Neuropsychologia," 10, 437-445. . KIMURA, D. (1966) Duel functional asymmetry of the brain in visual perception, "Nc'Jropsychologia," 4, 275-285. LEVY, J., TREVARTHEN, C., and SPERRY, R.W. (1972) Perception of bilateral chimeric figures following hemispheric disconnection, "Brain," 95, 61-78. McKEEVER, W.F. (1971) Lateral word recognition: Effects of unilateral and bilateral pre· sentation, asymmetry of bilateral presentation, and forced order of report, "Quart. J. Exp. Psychol., n 23, 410-416. - , and HULING, M.D. (1971a) 'Lateral · dominance in tachistoscopic word recognition performance obtained with simultaneous bilateral imput, "Neuropsychologia," 9, 15-20. - , (1971b) Bilateral tachistoscopic word recognition as a function of hemisphere stimulated and interhemispheric transfer time, "Neuropsychologia," 9, 281-288. - , SUBERI, M., and VAN DEVENTER, AD. (1972) Fixation control in tachistoscopic studies of laterality effects: Comments and data relevant to Hines' experiment, "Cortex," 8, 473-479.
Independent functioning of cerebral hemispheres
143
M.E. U973) Laterality differences in tachistoscopic word recognition in normal and delayed readers in elementary school, "Neuropsychologia," 10, 437-445. WHITE, M.J. (1972) Hemispheric asymmetries in tachistoscopic information processing, "Brit. J. Psychol.," 63, 497-508. (1973) Does cerebral dominance offer a sufficient explanation for laterality differences in tachistoscopic recognition, "Percept. Mot. Skills," 36, 474-485. OLSEN,
D. Hines, Ph. D., Department of Behavioral Science, The Milton S. Hershey Medical Center, Hershey, Pennsylvania
17033.
INTRODUCTION
Words are recognized more accurately from the right visual half-field (VHF) under conditions of tachistoscopic presentation. At least part of this superiority is often attributed to the more direct pathway between the right VHF and the language centers in the left hemisphere (Kimura, 1966; Hines, 1972b). This effect has usually been reported using stimuli presented unilaterally to only one VHF, on any given trial. Recently, however, McKeever (1971) has shown that when words are presented bilaterally to both VHF's, with fixation control using a center digit, a significantly greater right VHF superiority is produced. This effect is quite large. With unilateral presentation McKeever (1971) found a right VHF superiority of about 1.5 to 1; with bilateral presentation he found a ratio of about 5 to 1. The 5 to 1 ratio has also been reported by McKeever and Huling (1971a) and by Hines ( 1972a). Several hypothesis to explain the larger right VHF superiority with bilateral stimulation have been proposed and tested, but none have satisfactorily accounted for all the experimental data. McKeever (1971) tested the effect of order of report on the bilateral VHF asymmetry but he found the large right VHF asymmetry was maintained even when subjects were instructed to report the left VHF first. McKeever and Huling (1971b) tested a "temporal lag" hypothesis to explain the larger VHF asymmetry. They suggested that the right VHF word arrived first at the language areas in the left hemisphere, since it did not have to cross the callosum. They also hypothesized that the left VHF word arrived in the left hemisphere, via the callosum, while the left hemisphere language center already was "engaged in processing or recovering from processing" the right VHF word. McKeever and Huling tested this hypothesis by presenting the left VHF word either 6 msec., 10 msec., or 20 msec. prior to the right VHF word. The 6 msec. time was thought to Cortex (1975) 11, 132-143.
Independent functioning of. cerebral hemispheres
133
be roughly the time necessary to cross the callosum. However, they found that earlier presentation of the left WIF stimulus did not result in a decrease in right VHF superiority. Hines (1972a) and White (1973) suggested that the center digit might, in itself, alter fixation. Hines suggested that after reading the center digits, .. subjects may simply continue reading to the right." However, McKeever, Suberi and Van Deventer (1972), Olsen (1972) and Kershner and Jeng (1972) demonstrated that other methods of fixation control which do not use a center digit also produce the large right VHF superiority. The present experiment tests two additional hypotheses to explain the increased right VHF superiority with bilateral stimulation. The first hypothesis is that the increase is due to subjects directing attention toward the more easily recognizable right VHF word, and tending to ignore the left VHF word. This hypothesis will be tested in two ways. First, we will evaluate whether accurate report of words from the right VHF is associated with decreased accuracy from the left VHF on any given trial. Second. we will test whether the large right VHF superiority is obtained when words are presented in one VHF and nonverbal stimuli in the opposite VHF. With bilateral presentation of words, very few words are recognized from the left VHF. However, nonverbal shapes show much less asymmetry with bilateral presentation and fixation control. Hines (1972a) found that subjects reported about 25 % of the shapes from the left VHF with bilateral presentation but less than 10% of the words. If subjects can also recognize about 25% of the shapes from the left VHF when they are paired with words, without lowering recall of words from the right VHF, then it will suggest the left VHF "deficit" is not due to attentional factors, but rather is specific for verbal stimuli. If, however, subjects cannot recognize nonverbal stimuli from the left VHF, without decreasing recognition of words from the right VHF, it will suggest that attentional factors may be important. The second hypothesis to be tested is based on Dimond's (1972) "channel" theory of visual perception. Dimond postulates that stimuli to each VHF are received and analyzed by the contralateral hemisphere. Thus any stimuli presented to the right VHF will be initially processed by the left hemisphere, while stimuli to the left VHF will be initially recognized by the right hemisphere. The recognition accuracy from each VHF depends on the processing efficiency of its contralateral hemisphere for any particular type of stimuli. Dimond has also proposed that while bilaterally presented visual stimuli are received and recognized separately by the two hemispheres working as "independent channels," the stimuli may interfere with each other if they have to compete with each other for access to the expressive language centers located in the left hemisphere. Dimond's channel theory can be extended to account for the increased right VHF superiority with bilateral stimulation
134
D. Hines
as due to interference between the hemispheres since both must use the same expressive language areas. Thus the left hemisphere recognition channel may have preferential access to the language areas of the left hemisphere. In the present study, we would predict that subjects should show suppression of words from the left VHF due to interference only when words are presented bilaterally to each VHF. When the words are presented bilaterally, they may compete for the language mechanisms in the left hemisphere causing a loss of the word processed by the right hemisphere. However, when only a recognition of a nonverbal stimulus is required of the left hemisphere, then the verbal information from the right hemisphere should not be suppressed since the left hemisphere is not specialized for nonverbal recognition (Kimura, 1966; Levy, Trevarthen and Sperry, 1972). Thus when words are paired with nonverbal stimuli they should show a similar asymmetry to words presented unilaterally on successive trials. Three types of stimuli will be used in the present experiment: words, nonverbal shapes, and pictures of faces. A center digit will be used to control fixation. In all conditions, the stimuli will be presented bilaterally. In three conditions the same type of stimulus will be presented to each VHF: words to both VHF's, shapes to both VHF's, and faces to both VHF's. In two conditions a word will appear in one VHF and a nonverbal stimuli in the opposite VHF: words paired with shapes and words paired with fates. In these two conditions, the VHF in which the verbal and nonverbal stimuli appear will be randomly alternated between trials. The'attentional theory suggests that correct recognition from one VHF should be correlated with decreased recognition from the opposite VHF in all conditions. It also suggests that superior recall of verbal stimuli from the right VHF should decrease recall of verbal and nonverbal stimuli from the ignored left VHF. Dimond's theory suggests that stimuli should interfere only when they compete for a common response mechanism. Thus words should be reported more accurately when paired with nonverbal stimuli, particularly the words from the left VHF. MATERIAL AND METHOD
Subjects
Sixty three students and employees at the Hershey Medical Center of The Penn State University served as paid subjects. All subjects were naive as to the purpose of the experiment. Partial data from 18 subjects in the "shape-pairs" control condition have been reported previously (Hines, 1972a). Four additional subjects failed to make any correct responses and were excluded from the study. There were five experimental conditions in this study: words paired with words; words paited with. shapes; words paired with faces; shapes paired
Independent functioning of cerebral hemispheres
135
with shapes; and faces paired with faces. Fifteen subjects were randomly assigned to the words paired with shapes condition; fifteen subjects were assigned to the words paired with faces condition; fifteen subjects were assigned to the faces paired with faces condition and eighteen subjects were assigned to the shapes paired with shapes condition. Fifteen subjects received the words paired with words condition. Nine of these subjects had previously received the shapes paired with shapes condition and six the faces paired with faces condition. Equipment and stimulus preparation
The stimuli were presented using a Lafayette Model V-I constant illumination tachistoscope. The stimulus cards were prepared using index cards trimmed to fit the V-I tachistoscope. All cards had a number from one to nine typed at the central fixation point. Three types of stimuli were presented in the VHF's: words, shapes and pictures of faces. The words and central digits were typed on an IBM electric typewriter with standard pica type. The words were typed in all capital letters with one space between the letters and four spaces between the word and center digit. The inner edges of each word was 1.6 and the outer edge 4.1 degrees of visual arc from the center point of fixation at a viewing distance of 15 inches. Ten four-letter words were presented in the three word conditions (words paired with words, words paired with faces, words paired with shapes): CAKE, LANE, DOVE, MASK, EPIC, GOLD , HARE, FARM, POST and BEAR. All words appeared twice in each VHF in each of the three word conditions. The words used and stimulus preparation are identical to McKeever and Huling (1971b). The word and word pair stimuli were exposed for 20 msec. The "shapes" were prepared by the Medical Illustration Department of Penn State. They consist of inkblot-like shapes drawn to avoid resemblance to any familiar objects. On the stimulus cards, the inner edge of each shape was 1.9 degrees and the outer edge was 4.0 degrees of visual arc from fixation. Twenty of the shapes and a mirror image of each of the twenty were used as the stimuli in all shape conditions. Each of the shapes appeared once in one VHF and its mirror image appeared once in the opposite VHF in both the "shape and shape pairs" and "shape and word pairs" conditions. The mirror images were used to avoid any left-right recognition asymmetries inherent in the shapes themselves as suggested by Kimura (1966). For each trial an answer card containing six possible choices wasaIso prepared. Illustrative stimulus and answer cards have been published (Hines 1972a). The answer cards for each trial was placed adjacent to the tachistoscope. Subjects viewed the stimulus in the tachistoscope, then looked down at the answer card and pointed to the shape they felt they had seen. The shape and shape pairs and the shape and word pairs stimuli were exposed for 25 msec. A pilot study suggested that this was the shortest exposure at which most subjects could accurately recognize some shapes. The face pictures were obtained from photographs of medical students taken by the Department of Medical Illustrations at Penn State. The photographs were 3/4 inch by 7/8 inch in size and contained the full face of one student. The photographs were usually homogeneous with respect to background. darity, contrast, and face size. Twenty photographs and a mirror image of each photograph were utilized as stimuli in each face condition. There was 1.9 degrees of arc from the center to the inner edge of each photograph on the stimulus cards, and 4.5 degrees from fixation to the outer edge of each photograph. Each of the faces appeared once in one VHF on one trial and its mirror image appeared in the
136
D. Hines
opposite VHF on another trial. This procedure controls for any left or right bias inherent in the stimuli (Kimura, 1966). Forty answer cards were prepared for the "face and word" pairs conditions. Each card contained the correct face plus three alternatives on a 3" X 5" plain index card. The three alternatives were matched to the correct choice for sex, hair length, facial hair, presence or absence of glasses, dominant expression, background darkness, and orientation of the face. A pilot study suggested the latter two characteristics were particularly easy to discriminate in tachistoscopic exposure. The purpose of this matching was to eliminate differences between faces which could be easily verbalized. Twenty answer cards were also prepared for the face and face pairs condition. Those answer cards contained two correct choices plus three alternatives for each correct choice. The same alternatives for each face were used as in the face and word pairs answer cards. The appropriate card for each trial was placed adjacent to the tachistoscope. Subjects viewed the stimulus cards in the tachistoscope. then looked down at the answer cards and pointed to the face(s) they felt they had just seen. The face and face pair and the face and word pair stimuli were exposed for 30 msec. Pilot data suggested that this was the shortest exposure at which most subjects could accurately recognize some faces.
Procedure Sl,lbjects began each trial by viewing a card with 6 lines radiating from an open circle. The center circle was just large enough to "frame" the center digit. None of the 6 radiating lines was horizontal and they did not overlap with any VHF stimuli. Prior to each trial the subjects were alerted to ensure fixation onto the center circle. The stimulus card was then exposed, followed immediately by the return of the fixation card. Subjects in the word and shape pairs condition were instructed to fixate the center circle until the stimulus was exposed, to first report the center digit and then to name the word and select the shape from the answer card. There were forty trials for each subject. On each trial a word appeared in one VHF and a shape in the other. The VHF in which the word and shape appeared was randomly alternated between the trials. Subjects in the word and face pairs condition were instructed to fixate the center circle until the stimulus appeared, to first report the center digit and then to name the word and select the correct face from the answer card. There were forty trials for each subject. On each trial a word appeared in one VHF and a face in the other VHF. The VHF in which the word and face appeared was randomly alternated between trials. Subjects administered the face and face pairs condition were instructed to fixate the center circle until the stimulus appeared, to first report the center digit and then to select the two faces from the answer card. There were twenty trials for each subject. The faces in the two half-fields were dissimilar on at least one significant aspect of appearance. Subjects in the shape and shape pairs condition were instructed to fixate the center circle until the stimuli were flashed, then to report the number presented at fixation, and last to select the shape they had seen from an answer card. There were twenty trials for each subject. Subjects in the word and word pairs condition were instructed to fixate the center circle until the stimuli were exposed, report the number at fixation and
Independent functioning of cerebral hemispheres
137
then any words they recognized. This condition was essentially a bilateral replication of McKeever and Ruling (1971a). RESULTS
The mean correct responses for all conditions are shown in Table I. The data were first analyzed by analysis of variance to determine whether VHF asymmetry varied according to stimulus pairings for each type of stimuli. TABLE I
Mean Correct Score by Visual Half-Field for Each Condition Stimulus pairs
Mean correct recognition Left VHF
Right VHF
Mean Correct Words Words and Words Words and Shapes Words and Faces Shapes and Shapes Shapes and Words
2.4 4.7
9.9 7.8 10.3
Mean Correct Shapes 5.1 5.1
7.2 6.3
2.8
Mean Correct Faces Faces and Faces Faces and Words
5.6 5.5
4.8 6.2
The maximum possible score for each visual half-field is 20.
The mean correct responses for words was analyzed by a mixed analysis of variance with stimulus pairing as a between groups factor and the VHF's and interaction as within group variables. Significantly more words. were recalled from the right VHF in all stimulus pairings (F = 107.21; df = 1, 42; P < .001). However, the stimulus pairings did not significantly affect either overall recall (F = 1.28; df = 2, 42) or VHF asymmetry (F = 1.63; df = 2, 42). The mean correct response for shapes was also analyzed by a mixed analysis of variance. The right VHF was superior in overall recognition (F = 4.75; df = 1, 31; P < .05). Again, however, stimulus pairings did not alter either overall recall (F = .15; df = 1,31) or VHF asymmetry (F = .10; df = 1, 31). Mean correct recognition for faces was also analyzed using a mixed analysis of variance. The faces showed no significant difference between VHF's in overall recognition (F = .21; df = 1, 28). As with words and
D. Hines
138
shapes, stimulus pamngs did not alter either total recognitions (F = 1.45; df = 1, 28) or VHF asymmetry (F = .84; df = 1, 28). These analysis indicate that the different stimulus pairings did not significantly alter VHF asymmetry or overall recognition for either the words, shapes, or faces. The data were then analyzed to determine whether the verbal and nonverbal stimuli showed significantly different VHF asymmetries when paired together. Data from the shapes and words paired condition were analyzed by analysis of variance, with both VHF and stimulus type within group variables. The words showed a significantly greater right VHF superiority than did the shapes, resulting in a significant VHF X stimuli interaction W = 17.63; df = 1, 14; P < .001). Data from the word and face pair condition were similarly analyzed. Again the words showed a right VHF superiority significantly greater than the faces (F = 16.00; df = 1, 14; P < .005). Thus while stimulus pairing did not significantly alter VHF asymmetry, both types of nonverbal stimuli did show a significantly different ViHF asymmetry from the verbal stimuli. The data were finally analyzed to determine whether correct recall from one VHF was associated with incorrect recall from the opposite VHF on the same trial. For each subject, the number of trials in which both VHF's were reported correctly was compared with the number expected by chance simply on the basis of total recall in each VHF. For example, if in 20 trials a subject had 8 correct responses from one VHF and 10 correct responses from the other VHF, by chance 10 X 8/20 or 4 trials should have occurred in which both responses were correct. The mean number of trials per subject in which both VHF's were reported correctly.and the mean number of trials per subject expected by chance are presented for each stimulus condition in Table II. Differences between the actual and chance distribution were tested by the Wilcoxon signed-rank test. TABLE II
Mean Trials in which Stimuli are Correctly Reported from both Visual Hall-Field (VHF) as Expected by Chance Type of stimuli Right VHF
Left VHF
Words Words Shapes Shapes Words Faces Faces
Words Shapes Words Shapes Faces Words Faces
* p.OI ** p.02
Both stimuli reported correctly Change
Observed
1.87 .97 2.04 2.08 1.39 3.04 1.27
1.86 1.27 2.00 1.00* . 1.20 ·3.20 .73**
Independent functioning of cerebral hemispheres
139
In two conditions, significantly fewer double correct responses occurred than would be expected by chance. When shapes were paired with shapes less than half the expected double correct responses occurred (T = 16, N = 18, P < .01). When faces were paired with faces there were also less trials when faces from both VHF's were correctly reported than would be expected by chance. (T = 19.5, N = 15, P < .02). Significantly more double responses occurred when the nonverbal stimuli were presented to the left VHF and the words to the right VHF, than when words were presented to the left VHF and shapes or faces to the right VHF (shapes: T = 3, N = 8, p < .05; faces: T = 0, N = 11, P < .01). However this reflected higher overall recognition when the verbal stimulus was presented to the right VHF and the nonverbal stimulus to the left VHF and the number of double responses for each of the verbal-nonverbal pairing was not different from chance expectancy. DISCUSSION
The results of this experiment show that the overall recogmtlon and VHF asymmetry depended on the type of stimulus and not on the stimulus paidngs. The words showed a large right VHF superiority whether paired with words, shapes, or faces. The shapes showed a small but statistically reliable right VHF superiority whether paired with words or shapes. The faces showed no significant differences between VHF's whether paired with words or faces. The large right VHF superiority for words is typical of this bilateral recognition paradigm. The right VHF superiority for shapes is not typically found with unilaterally presented nonverbal stimuli. In 25 experiments summarized by White (1972), nonverbal stimuli were recognized more accurately from the left VHF in 7 experiments, no difference was found in 15 experiments, and a right WfF superiority was reported in only 3 experiments. The lack of asymmetry for faces is consistent with the unilateral findings. These results indicate that the large right VHF superiority for bilaterally presented words is not a result of attending the right VHF and ignoring the left VHF. If subjects preferentially attend one VHF, then correct recognition from that VHF should be correlated with impaired recognition from the opposite VHF. However the recognition accuracy from one VHF did not affect recognition accuracy from the opposite VHF on the same trial in any of the word conditions . .In addition, if the left VHF is simply being ignored, then all types of stimuli should be recognized with similar impaired accuracy when words are being recoghized accurately from the left VHF. However, the nonverbal stimuli could be recognized accurately from the left VHF while words were recognized with unimpaired accuracy from the right VHF. Finally,
140
D. Hines
even though words were recognized with low accuracy from the left VHF in all conditions, the high accuracy from the right VHF was only observed in the words and word pairs conditions. However, if subjects were preferentially attending the right VHF, it should increase recognition for all right VHF stimuli. The latter effect is particularly noticeable in the word and shape pairs condition. The data are also inconsistent with the hypothesis that the bilateral verbal stimuli interfere with each other because both use the same language' response mechanism. If this hypothesis were correct, then the large VHF asymmetry for words should only be observed in the word and word pairs condition, when both words must be reported using the left hemisphere expressive language areas. However, a large right VHF superiority for words was observed in all conditions. The different stimulus pairings did not significantly affect either VHF asymmetry or overall recognition for words. Also, if the interference hypothesis were correct, then correct recognitions from one VHF should be correlated with incorrect recognitions from the opposite VHF on the same trial in the word and word pairs condition. However, VHF recognition on any given trial were not correlated in the word and word pairs condition. The most interesting positive finding was the interference noted with the two bilateral nonverbal conditions. When a nonverbal stimulus was recognized from either VHF, it significantly lowered the probability of correct recognition of a nonverbal stimulus from the opposite VHF. This was found in both the face and face pairs and the shape and shape pairs conditions. This finding suggests that subjects are relatively limited in their ability to keep unfamiliar nonverbal stimuli in short term memory, perhaps to only one stimulus at a time. While the data from this experiment do not support Dimond's interference hypothesis for verbal stimuli, they are otherwise consisted with channel theory. Each type of stimuli displayed the same VHF asymmetry regardless of stimulus pairing. This is consistent with the theory that the response asymmetry is due to differences between the two hemispheres in recognizing different types of stimuli. A proposed new
selective channel theory
The results of this experiment are most consistent with the theory that the two hemispheres function as separate channels for recognizing visual stimuli only under conditions of bilateral tachistoscopic VHF stimulation with fixation control. In this paradigm each hemisphere serves as an independent channel for recognizing stimuli from its contralateral VHF. VHF asymmetries under this condition reflect actual differences between the hemispheres for recognizing different types of stimuli. Thus the large right
Independent functioning of cerebral hemispheres
141
VHF superiority with bilateral presentation reflects the superior verbal recognition ability of the left hemisphere. It is also hypothesized that the two hemispheres do not function as separate channels under unilateral VHF tachistoscopic presentation. Rather, with unilateral VHF presentation both hemispheres contribute to recognizing incoming stimuli, with the primary role played by the hemispheres specialized for processing any given stimulus. Thus VHF asymmetry for tachistoscopically presented unilateral stimuli do not directly reflect differences between the two hemispheres in recognition ability. Rather it is due to information loss in transmission to the most efficient hemisphere when stimuli must cross the callosum. Thus the degree of VHF asymmetry with unilateral stimuli reflects the extent to which they are vulnerable to a slight loss of fidelity. Small complex stimuli (such as words) which are presented at rapid rates should be most affected. Larger stimuli presented at slower rates should be less affected and show less asymmetry. Selective channel theory thus suggests that differences in VHF asymmetry for words between the unilateral and bilateral conditions reflect two distinct and different mechanisms. The large right VHF superiority for words in the bilateral condition directly reflects the superior ability of the left hemisphere to recognize verbal stimuli. The smaller right VHF asymmetry with unilateral stimulation reflects a loss of information in transmission to the left hemisphere across the callosum. Selective channel theory finally proposes that the two hemispheres work as independent channels only when discrete tachistoscopic stimuli are presented bilaterally to each VHF, with positive fixation control. Under other bilateral presentation conditions, verbal stimuli are recognized sequentially from left to right with the recognition performed primarily by the left hemisphere. This, of course, is how English is normally read. The left to right order of recognition has also been found when letter spans are presented tachistoscopically across fixation to both VHF's (Heron, 1957; Bryden, 1960). Subjects report that they actually attend each letter from left to right (Heron, 1957). However, when fixation is controlled by a center digit, the left-to-right scan is no longer possible. Under these conditions, the two hemispheres work as independent channels to analyze information from their contralateral VHF.
SUMMARY
Bilateral tachistoscopic presentation of verbal stimuli produces a significantly larger right visual half-field (VHF) superiority than does unilateral presentation, when fixation · is controlled by a center digit. This experiment tested whether the increased asymmetry was due to (a) subjects attending the right VHF and ignoring the left VHF or (b) interference between the hemispheres due to competition for the left hemisphere language areas.
142
D. Hines
Words, shapes, and pictures of faces were presented bilaterally to each VHF, with fixation controlled by a center digit. In three conditions, the same type of stimuli was presented in each VHF (e.g., a word in both the left and right VHF). In two conditions words were presented to one VHF and nonverbal stimuli to the opposite VHF (words paired with words and words paired with faces). It was found that the stimulus pairings did not affect VHF asymmetry for any stimulus. Words showed a large right VHF superiority in all conditions. Shapes showed a significantly smaller right VHF superiority in all conditions. Faces showed no VHF asymmetry in any condition. It was concluded that attentional factors were not important ~ince shapes or faces could be recognized accurately from the left VHF without lowering verbal recognition from the right VHF. Thus the low recognition accuracy from the left VHF is specific for verbal stimuli rather than attentional. The interference hypothesis was also not supported since all the right VHF stimuli (words, shapes, or faces} were associated with low recognition of words from the left VHF. It was suggested instead that VHF asymmetry under unilateral and bilateral presentation reflect two different mechanisms. Under conditions of unilateral presentation, VHF asymmetries are caused by loss of information when any given stimulus must cross the callosum to reach the hemisphere specialized for its processing. However, with bilateral VHF presentation and fixation control, the two hemispheres act as independent channels for information processing. Under this condition, each hemisphere recognized the stimulus from its contralateral VHF. Thus the large .right VHF superiority for words with bilateral presentation reflects the superior ability of the .left hemisphere for verbal recognition.
REFERENCES BRYDEN, M.P. (1960) Tachistoscopic recognition of non-alphabetical material, "Canad. J. Psycho!.," 14, 78-86. DIMOND, S. (1972) The Double Brain, Williams and Williams, Baltimore. HERON, W. (1957) Perception as a function of retinal focus and attention, "Am. J. Psycho!.," 70, 38-48. HINES, D. (1972a) Bilateral tachistoscopic recognition of verbal and nonverbal stimuli, "Cortex," 8, 315-322. . ' - (1972b) A brief reply to McKeever, Suberi and Van Deventer's comment on "Bilateral tachistoscopic recognition of verbal and .nonverbal stimuli," "Cortex," 8, 480-482. KERSHNER, J.R., and JENG, A.G. (1972) Duel functional hemisphere asymmetry in visual perception: Effect of ocular dominance and postexposure processes, "Neuropsychologia," 10, 437-445. . KIMURA, D. (1966) Duel functional asymmetry of the brain in visual perception, "Nc'Jropsychologia," 4, 275-285. LEVY, J., TREVARTHEN, C., and SPERRY, R.W. (1972) Perception of bilateral chimeric figures following hemispheric disconnection, "Brain," 95, 61-78. McKEEVER, W.F. (1971) Lateral word recognition: Effects of unilateral and bilateral pre· sentation, asymmetry of bilateral presentation, and forced order of report, "Quart. J. Exp. Psychol., n 23, 410-416. - , and HULING, M.D. (1971a) 'Lateral · dominance in tachistoscopic word recognition performance obtained with simultaneous bilateral imput, "Neuropsychologia," 9, 15-20. - , (1971b) Bilateral tachistoscopic word recognition as a function of hemisphere stimulated and interhemispheric transfer time, "Neuropsychologia," 9, 281-288. - , SUBERI, M., and VAN DEVENTER, AD. (1972) Fixation control in tachistoscopic studies of laterality effects: Comments and data relevant to Hines' experiment, "Cortex," 8, 473-479.
Independent functioning of cerebral hemispheres
143
M.E. U973) Laterality differences in tachistoscopic word recognition in normal and delayed readers in elementary school, "Neuropsychologia," 10, 437-445. WHITE, M.J. (1972) Hemispheric asymmetries in tachistoscopic information processing, "Brit. J. Psychol.," 63, 497-508. (1973) Does cerebral dominance offer a sufficient explanation for laterality differences in tachistoscopic recognition, "Percept. Mot. Skills," 36, 474-485. OLSEN,
D. Hines, Ph. D., Department of Behavioral Science, The Milton S. Hershey Medical Center, Hershey, Pennsylvania
17033.